Although myotonic dystrophy (DM) is a multi-systemic neuromuscular disorder, previous mechanistic studies have emphasized disease-associated heart and skeletal muscle abnormalities. However, the DM brain is also profoundly affected with abnormal behavior and cognition, progressive memory loss and cerebral atrophy. The long-term objective of the proposed research is to elucidate the molecular basis for CMS pathology in DM. DM is an RNA-mediated disease caused by the expansion of CTG and CCTG microsatellites in two different genes. Following transcription, mutant CUG and CCUG expansions accumulate in the nucleus and inhibit the RNA alternative splicing activity of the muscleblind-like (MBNL) proteins. During normal postnatal development, MBNL proteins promote skipping of specific fetal exons during RNA splicing by antagonizing the CUGBP1 and ETR-3-like splicing factor (CELF) proteins. Thus, DM is caused by retention of fetal protein isoforms in the adult. While this model explains specific aspects of muscle pathology, including myotonia, it is unclear if a similar pathogenesis mechanism occurs in the brain. This proposal is divided into three specific aims designed to pursue the hypothesis that specific defects in gene expression and RNA splicing induced by expansion RNAs results in the CNS phenotype associated with DM. First, we determine if loss of Mbnh gene, or poly(CCUG) transgene, expression results in similar DM-relevant alterations in the expression and splicing patterns of specific gene transcripts in several different regions of the developing and postnatal brain. Second, we will test the hypothesis that conditional loss of MbnM and/or Mbnl2 gene expression in the brain results in postnatal neurological deficits while MbnIS expression is essential for normal embryonic brain development. Third, recombinant adeno-associated virus (rAAV)-mediated transduction of neurons will be used to test the possibility that abnormal RNA splicing and expression patterns in transgenic poly(CCUG) mouse models for DM1 and DM2 can be modulated by overexpression of Celf and Mbnl proteins in vivo. These studies should elucidate the molecular events underlying DM pathogenesis in the CNS and identify molecular targets for future therapies.